Skip to main content
European Commission logo print header

Scientific Approach to Finding Indicators of and REsponses to Radicalisation

Final Report Summary - SAFIRE (Scientific Approach to Finding Indicators of and REsponses to Radicalisation)

Executive Summary:
Radicalisation is not a new phenomenon. Nor is it uncommon. Some would argue that we are all radicals at some level. Radicalisation that leads to violence, however, is relatively rare but can have a huge impact on society: financially, emotionally and in terms of the security of innocent citizens. The key in democratic societies is to ensure citizens’ rights to free thought – even radical thought – while protecting society from the fallout of illegal actions from violent radicalised groups and individuals. Successfully achieving this goal depends on understanding the phenomenon of radicalisation from its roots in thought and discourse to the stage where individuals go beyond it and engage in violent and illegal behaviour in the name of their cause.

SAFIRE: Scientific Approach to Finding Indicators of & REsponses to Radicalisation. SAFIRE is an EU project started in June 2010 to explore this interesting and sensitive topic. The scope of SAFIRE primarily involves groups and individuals on the extreme and violent end of the radicalisation spectrum. However, in order to understand them and their motives, we also need to step back and understand what happened before they turned to a more violent version of their philosophy. In this project, we focus on two innovations in this field of research:
• Developing a non-linear model of the radicalisation process based on typologies of radical groups, cultural aspects of radicalisation, observable indicators of radicalisation, interventions designed to reverse, halt or prevent the radicalisation process.
• The collection of qualitative and quantitative empirical data to test hypotheses about radicalisation and principles of effective interventions.

We developed and carried out the SAFIRE research with the explicit application to policy and field in mind. Some relevant practical results are:
• Intervening in the pre-violent stage of radicalisation is not supported by all EU Member States. SAFIRE examines the arguments on both sides of the discussion. Pre-violent intervention may be ethically justifiable if it helps adolescents to be better able to critically think about their life and to decide for themselves what is right for them.
• Various cultural factors seem to make a society more or less susceptible to radicalisation. For example: national wealth, state (in)stability, integration policies, political participation or trust in authorities.
• Part of the challenge of dealing with violent radicalisation is separating the wheat from the chaff: how can we identify key determinants of radicalisation in the enormous mass of seemingly relevant factors? SAFIRE demonstrates that the science of network studies can help. It can help guide the choice of interventions, identify which factors are central and which are peripheral to a particular aspect of radicalisation, and help decide which factors can be targeted when trying to prevent radicalisation-related violent acts.
• Different types of radicalised groups have different characteristics and, as such, require a different approach to deal effectively with the threat of violence they pose to society. One, single, homogeneous approach lacks the nuance to deal with the complexity of the problem. In SAFIRE we identified common and distinguishing characteristics of radicalised groups. This work shows that there are various ways to categorise radicalised groups and that traditional ideological categorisations are not always sufficient to capture group differences.
• The Internet is an increasingly important environment for individuals on the way towards radicalisation. There is a challenge for law enforcement as it facilitates and enhances the opportunities to become radicalised and to spread violent extremist beliefs. The Internet also provides a mechanism through which to move towards violent extremist or terrorist acts in ways that are difficult to detect and easy to anonymise. It also has advantages: European governments may wish to make use of the internet to determine and propagate strategies, policies, and resource allocation.
• Indicators of radicalisation cannot be sought in unchanging behaviour or in any one individual behaviour. Instead, identifying high-risk individuals or groups should be sought in the observation of certain changes that occur in concert over time.
• Most successful interventions designed to prevent violent radicalisation tend to focus on psychological factors, such as self-esteem, dealing with negative emotions, and reducing feelings of injustice. In addition, offering alternatives (education, accessibility to internships) and developing personal skills (conflict management) help reduce susceptibility to violent radicalisation. Interventions at a more advanced stage of radicalisation should focus on individuals and be carried out in close cooperation with people who are credible in the eyes of the targeted individual (i.e. former members of extremist groups).

For more information on SAFIRE and the project’s results, we refer you to the various focus documents on our results website:
Project Context and Objectives:
1.1 Summary of project objectives
The goal of SAFIRE is to improve fundamental understanding of radicalisation processes and use this knowledge to develop principles to improve (the implementation) of interventions designed to prevent, halt and reverse radicalisation. SAFIRE developed a process model of radicalisation, giving insight into the process from moderation to extremism, based on a non-linear dynamic systems approach and a typology of radical groups. This is an innovative approach that has not been explicitly applied to this area up until now. Principles regarding interventions were developed in close concert with the models, and were applied in a longitudinal, empirical study. Important aspects of radicalisation that also addressed were: the relationship between national culture and radicalisation, radicalisation on the Internet, and defining observable indicators of the radicalisation process. The main deliverable (the website contains on one hand a detailed description of the work done in SAFIRE and on the other an interactive, accessible overview of the project designed for quick, user-friendly accessibility to the main points. Envisaged end-users are policy makers, researchers in the field of radicalisation and professionals who work with high-risk individuals. The results of this project increase the understanding of both conceptual aspects of radicalisation, and practical characteristics and modus operandi of radical groups. In addition, the results increase understanding of field efforts and interventions – when, why and how they work – thus helping focus the allocation of resources and the implementation of interventions.

1.2 Description of the work performed since the beginning of the project
The goal of the proposed project was twofold. The first goal was to increase scientific insight into the process of radicalisation. The second goal is to provide theoretical argumentation and empirical evidence for the implementation of practical interventions and related means. The project was made up of four steps. In the first 14 months of the project, we completed Step 1. Between month 15 and 28, we carried out Steps 2 and 3 and in the final period we devoted our effort to Step 4.

In Step 1 of SAFIRE, we conducted a broadly-oriented and extensive inventory to provide the basis needed to extend the state of the art. We examined known radical groups and their characteristics – in the broadest sense of the word – thus providing a comprehensive overview. Using analysis of information on the Internet we developed an overview of the scope (both in terms of quality and quantity) of radicalisation on the internet. In an inventory of field interventions, we developed an overview of what is being done to deal with the threat of radicalisation in countries both inside and outside the EU.

The conceptual work in Step 2 refers to 1) the development of a model describing the process of radicalisation using principles of dynamic, non-linear systems, and 2) an analysis of the role of national culture in this process. Important in this step was to identify behaviours and - more importantly - links between characteristics and behaviours that typify radicalisation. The information was always traceable and we strove towards maximising the verifiability and reliability of the information.

Step 3 turned to the field. In SAFIRE, we evaluated the effectiveness of interventions and identified and further developed observable indicators of radicalisation. These were linked to the in the process model; work important to understanding the nature of involvement in violent radicalisation and allocating resources to combat it (Horgan, 2008).

In Step 4, we set up and conducted a scientifically-designed, longitudinal empirical study in which we operationalised the concepts and studied them in a real-life setting for a finite period of time. The aim of the study was to evaluate the process model and related concepts, examine the intervention, and demonstrate ways in which these can be implemented in, and be of use to the field.

1.3 Description of main results
The main results include:
• From the background work in WP2
o All terrorists have been radicalized but not all radicalized individuals go on to participate in terrorism
o There is no single profile of a terrorist and no single entry pathway through the process of radicalization; the process manifests itself in a complex and highly individual-specific way.
o No single factor, or combination of factors, is common to radicalisation associated with different terrorist ideologies e.g. religious, political or nationalist-separatist
o Despite this, it may be possible to use similar methods in intervention programmes aimed against individuals with different types of motivation.
o Many intervention programmes are described in the literature but there is no firm evidence to show the causality of any results claimed by the programmes, nor does the literature identify, evaluate or compare intervention programme ‘best practices’ or ‘lessons learned’.
• Observable indicators
o Changes in the individual or the group are key for identifying potential radicalisation
o Changes in multiple factors are a pre-requisite for radicalisation
o Association with a charismatic radical leader is always a risk
o 21 indicators are divided into five categories:
 Self-identification
 Us vs. Them Societal View
 Social Interaction
 Persona
 Association
o Beware of false positives
• Evaluation of interventions
o Interventions rated as most effective focus on creating positive identity and reducing negative emotions
o Interventions rated as least effective focus on restoring authority
o Interventions were rated as more effective when they were given by former radicals. This may have to do with increased trust and demonstrated achievability, relative to people who have not experienced being a radical first hand.
• Modelling work
o Descriptive modelling work in WP4.1 shows radical groups and individuals can be described by characteristics other than their ideology. In addition:
 Nationalist-separatist groups and left-wing groups share modus operandi, hierarchic structure and recruitment methods
 There are multiple types of Islamist radical groups that differ in command and control structure, geographical scope of their actions, capacity to conduct successful actions, and methods
o Radicalisation is also described as a multi-disciplinary network
 Factors influencing radicalisation can be identified at the individual, group and societal level
 Some are actionable, some are not
 Influencing radicalisation can be realised by influencing the actionable factors
• Culture and radicalisation
o Three categories were distinguished: Socio-demographic and economic factors; Political and institutional factors; Psychosocial and cultural factors
o Socio-demographic and economic factors, such as wealth, seemed to have the strongest link with radicalisation

1.4 Final results and potential impact and use
The results of this project help increase the understanding of conceptual aspects of radicalisation, such as the psycho-social dynamics of radical groups and individuals, and practical characteristics and modus operandi of radical groups. In addition, the results increase understanding of field efforts and interventions – when, why and how they work – thus helping focus the allocation of resources and the implementation of interventions.

In the final product, we provide the disseminable deliverables, accompanied by 23 focus documents that highlight the most important and notable results and insights from the project, both conceptual and practical, and extend these by placing them in an integrated context.

Envisaged end-users are policy makers at the local, regional, national and European levels, who among other things are responsible for setting research agendas; scientists engaged in research on radicalisation; and professionals who work with high-risk individuals (police, social workers, educators, etc.). The SAFIRE project included an Advisory Board made up of experts from all these end-user groups.

Impact is two-fold. First there is the impact of the work itself. This is somewhat limited due to the restricted nature of two of the most important deliverables; only public summaries of the work will be freely disseminated, which prohibits among other things, dissemination in scientific journals. Second there is the impact this project has on setting precedents for conducting research on radicalisation – a topic that is both scientifically and ethically difficult to study. D1.3 addresses this point in detail and provides direction for such research in general, and EU projects in particular, in the future.

1.5 Project website information
SAFIRE has a public and a password-protected private website: and respectively. The content of the public website consists of general information on the objectives, main deliverables and consortium partners of the SAFIRE project. On the private website consortium partners share relevant documents regarding the different (sub)work packages. There are also an LinkedIn Discussion group and a e-newsletter tool, as part of the public website. The project results are showcased on a dedicated website:
Project Results:
For all figures and tables please see the attachment Figures and Tables.pdf

The meaning of the SAFIRE acronym is Scientific Approach to Finding Indicators of and Responses to Radicalisation. This is a fairly good framework with which to think about the results of the project: Scientific – Approach – Indicators – Responses – Radicalisation. For describing the results, we will torture the acronym a bit to: FARIRS (Finding – Approach – Radicalisation – Indicators – Responses – Scientific), and use this to structure the presentation of the foreground of the project.
Here, Finding refers to how we went about our work. What methods did we use to study radicalisation? Approach refers to the ethical boundaries of the project. As radicalisation is both a morally and politically sensitive topic, it was of the utmost importance in SAFIRE that we treated the topic with the care it deserves. Radicalisation refers to how we scoped the problem. How did we describe radicalisation and its context? Indicators refers to the work we did on observing radicalisation: can it be observed, and if so, what should you look for? Responses refers to intervening in the radicalisation process. What types of interventions are out there? Which interventions work best? Should we intervene in the process of radicalisation? Finally, Scientific refers to the empirical work we did examining the effects of interventions and improving our understanding of the radicalisation process.
In the beginning of this project, the main goal was to develop a non-linear model describing the process of radicalisation, from moderation to violence. Though we did indeed do this, at the end of the day, the project has yielded so much more; the whole of the results is much more than the sum of the parts. We have a better understanding of

• what types of radical groups there are and what characterises them
• why people join and leave radical groups
• what are determinants of radicalisation and how do they relate to each other
• how to do research on radicalisation and with (potentially) radical individuals
• how and why to intervene in the radicalisation process
• what types of interventions work best
• how is radicalisation related to culture
• what are observable indicators of radicalisation
• what are the ethical considerations and constraints to intervening in the radicalisation process on one hand and doing work on radicalisation on the other.

In this section of the final report, we will discuss these topics in light of the findings of the SAFIRE project.


In SAFIRE we used multiple methods to examine radicalisation. In doing the background study, we used literature review and an analysis of radicalisation as it is found on the Internet (media analysis). To study interventions and typologies we augmented the literature review with interviews and focus groups. Interviews were also used to gain insight into the process of radicalisation, specifically why people join radical groups and why they leave. In a longitudinal field study we were able to follow participants in a preventive intervention programme designed to help foster self-efficacy and resilience, thereby reducing their risk of joining radical groups. Finally, we used social network analysis to develop a multi-level model of the determinants of radicalisation.


From the very start of the project, ethical considerations regarding radicalisation were central to our work. Using ethical parallel research (WP 1.2) we monitored the ethical issues in the project as they came up and decided on the best course of action. Here we identify some of the most pressing issues that came up during the project.

The discussion of non-violent radicalisation

Different countries in Europe think about radicalisation and the threat it poses to our societies in different ways. The most relevant difference in the context of SAFIRE is the permissibility of the discussion of non-violent radicalisation. You can think about radicalisation as a continuum ranging from non-violent, but extreme ideas on one end to full-blown terrorism on the other. In some countries such as France and Portugal, the right to free speech and thought far outweigh the potential threat that non-violent radicals. Discussion of the non-violent end of the spectrum are out of the questions: if there is no evidence of illegal activity, individuals and groups have the inalienable right to think and say what they like. Moreover, when there is evidence of illegal activity, it should be addresses in terms of law-breaking rather than in terms of radicalisation, ideology or ethnicity.
In other countries, such as The Netherlands and Germany, it is permissible and social acceptable to discuss non-violent radicalisation and the threat it may pose to society and security. Interventions designed to prevent (further) radicalisation are acceptable. Policy is focussed on preventing rather than curing. Here, radicalisation is seen not only in terms of (il)legality, but rather as something that is immoral and unacceptable due to the potential impact to society should it become violent. Addressing it early on in the continuum reduces the risk of serious problems later on.
It is not for us to say that one approach is better or worse than the other. However, we do note that such disparity hampers discussion and eventual harmonisation of policy and field efforts at the international level.

To intervene or not to intervene?

Even in societies in which it is acceptable to discuss non-violent radicalisation, it is clear that you cannot just intervene whenever you want with whomever you want. Especially when interventions are geared towards high-risk individuals who have actually broken any laws or engaged in radical activities, interventions should be undertaken with care. In the ethical research in SAFIRE, we considered if ethical theories can provide justification for preventive intervening in the (pre-violent) stages of radicalisation.
In virtue ethics, the theory argues that everyone has the right to live a good life, though what ‘the good life’ is, is not the same for everyone; each person has to discover what the good life is for them. To do this, two things are pre-requisite. First, the individual has to be able to critically consider and decide what they want for themselves. To do this, they have to develop the skills necessary to be able to reason, critically reflect and discuss ideas with others. Second, there have to be other humans with whom the individual can discuss and refine their ideas. If one’s group – be it either radical or otherwise – does not allow for critical discussion, individual thought and individual differences, then being a member is by definition not conducive to achieving the good life. Thus, intervention programmes that focus on helping individuals develop the skills necessary to discovering their own good life, are ethically acceptable.
Two notes. First, the benefit of such a programme are not limited to (potential) radicals. Everyone, particularly during adolescence can benefit from interventions that help them discover what they want from life. Second, virtue ethics is clear on the fact that there is no one good life: it is different for everyone. Therefore, if an individual concludes that being a member of a radical group is what is good for them, then that is ethically acceptable.

Research ethics

When doing radicalisation research, it is important to consider that you will tend to work with vulnerable individuals. In addition, mainstream society can put a considerable stigma on being a radical. For these reasons, informed consent and anonymity are of particular importance in this type of research. Often, we found that participants did not want their identity to in any way be associated with the project. For this reason we had to develop an indirect informed consent method, in which informed consent is given via a third party. Anonymity should be protected at all costs: extra care should be taken when handling and storing data, participants’ names should not be used for example when making appointments etc.


Clusters of terrorist groups and individuals

We examined the literature on factors associated with radicalisation leading to violence or terrorism (WP 2.1). From this we categorised six types of radical groups: Islamist, right-wing, left-wing, nationalist-separatist, single issue (such as animal rights activists) and new religious movement (formerly referred to as cults). Notably, we found no unifying ‘terrorist profile’: every violently radical individual and group has their own story, their own path they have travelled to becoming radical. Though it is not possible to define a unifying path to terrorism, we can identify from the literature a number of factors common to the radicalisation of known terrorists. We identified 35 factors from the literature. The most common of these are:

• Political activism
• Inability to affect political change. Note that this feels intuitively contrary to the characteristic, political activism. However, it need not be. Attempting to enact political change through participation in the legitimate political system, combined with continued failure to do so effectively, may lead individuals to turn to alternative mechanisms.
• Indoctrination
• Proneness to violence
• Experience of negative political and personal events. This refers to experiencing specific (negative) events in one’s life that can trigger, or at the very least encourage, turning to radical means.

Another way to look at commonalities among radical individuals and groups is to consider common characteristics. To do this we asked the question if it were possible to describe terrorist groups in terms other than their ideology (WP 4.1). To do this, we analysed open information on known radical groups and individuals. This yielded five types of groups and five types of individuals, which can be described on multiple dimensions. For example, for the groups these were: command and control structure, tangibility of claims, leadership structure, resilience level, support base, weaponry. This yielded an overview such as presented in Figure 1. Figure 2 shows the clusters for individuals.
Overall, from this analysis we can conclude that not all terrorist groups or individuals are alike. Here, we will focus on groups. For more information on individuals, please see D4.1. For groups territorial aspects are very important in characterising a group. In some cases the clusters of groups cut across ideology suggesting that ideology is not the only defining characteristic. In this way it is worth noting that Islamist groups are distributed among three clusters, which suggests it is too short-sighted to think about all Islamist groups in a homogeneous way. Furthermore, different clusters will not always benefit from the same intervention: different clusters, different interventions. However, because different ideological groups may be found in the same cluster, they may benefit from the same approach.
In sum, it does seem that there is more to terrorists than their ideology. Courses of action to deal with the various types of groups can be found in approaches at the political, law enforcement and extra-judicial level.

Network modelling

One of the main goals of SAFIRE was to define the problem space of radicalisation: what are the main factors or determinants of radicalisation, what are the relevant contextual factors? And then of course to address the ensuing question of how to influence the process. To do this, we developed a relational model based on social network analysis (WP 4.1). This model included factors from different disciplines: psychology, political science, cultural psychology, cultural anthropology, criminology, and intervention practice. Information from these disciplines reflects different contextual levels of the problem space: individuals, groups and societies.
Figure 3 shows an overview of the model. For present purposes, it is not essential to be able to read the individual labels of each node in the model. What is apparent is that each discipline has its own cluster (criminology, for example, is the red cluster). This is partly due to the use of jargon specific to each discipline, which hampers defining overlap between disciplines. This is not only important in the academic exercise of building a model. Using the same terms for the same concepts is important to facilitating mutual understanding and cooperation between players in the field of radicalisation. Also notable is that, though the disciplines are mainly seen to be islands, there are bridges between them: concepts that are relevant to both disciplines.
Figure 4 shows an excerpt from the model describing the primary and secondary factors that influence an individual’s desire to gain membership in a radical group. Figure 4 shows gaining membership on the right. The column of factors just to the left are the factors that directly affect gaining membership, such as because you are searching for meaning, which you believe the group may provide. The column on the far left shows secondary factors related to gaining membership. The key is that by addressing factors that are related to the thing you want to influence, you are most likely to succeed. Overviews such as in Figure 4 show not only what factors are relevant, but also help the user identify which factors they can influence and which they cannot.


When thinking about indicators of radicalisation, we consider how to identify a risk of violent radicalisation at the macro level and at the micro level. The macro level refers to the level of national culture (WP 4.2). Are some national cultures more susceptible to radicalisation than others? Can we define characteristics common to high-risk cultures? On the micro level we looked at the possibility of defining observable indictors at the individual level to ascertain if someone may be radicalising (WP 3.2)

Cultural factors

Characteristics of a culture are not something you can change; they are embedded in that society. However, understanding cultural factors can be useful for understanding processes going on in that culture; they provide context and aid in interpretation of what you can observe. With radicalisation, we can also look for cultural characteristics more or less common to societies that have a high risk of violent radicalisation. We examined scientific literature and European databases with information relevant to our search for common cultural factors. The factors we found fall into five categories: socio-demographic and economic, political and institutional, psychosocial and cultural (composites of psychological and cultural characteristics at the country level), potential radicalisation (variables that reflect polarised attitudes to social objects), and indicators of violent radical behaviour.
Our results show first of all that there is very little explicit consideration of the role of culture in the process of radicalisation and terrorism. Having said that, we did find some evidence for cultural characteristics linked with the radicalisation and terrorism.

• Socioeconomic and political variables (GDP, income inequality, unemployment)
• Political and institutional factors (state instability, integration policies)
• Cultural history of violence
• National differences in willingness to participate in the political process
• Uncertainty avoidance
• Lack of trust in authorities

Note that radicalisation and terrorism are complex and diverse and cannot be attributed to a single cause. Nevertheless, combinations of cultural factors, specifically when certain political and economic conditions are present, can provide a fertile ground for violent radicalisation to emerge.

Individual factors

We asked ourselves if it were possible to actually recognise someone who is in the process of radicalising. Accepting that were are unable to look into someone’s thoughts, are there outward common characteristics indicative of radicalisation? Frontline workers, who had worked with radicals, helped us answer this question in focus groups. Before presenting the indicators, three notes:

• An isolated indicator is not informative. Multiple indicators should be observed and interpreted in light of the individual’s context before being able to conclude that someone is radicalising. The only indicator that needs attention when appearing in isolation is association with a known violent radical leader.
• Change is key. Indicators are only informative if they denote change from the way the individual used to behave.
• You cannot make conclusions without context. All changes observed need to be interpreted in light of an individual’s specific context.

We found five types of observable indicators relevant to ascertaining if radicalisation is occurring. Table 1 sums them us, together with specific manifestations of the five types of indicators. Note that more manifestations are possible.


The discussion of whether or not it is acceptable to intervene in the radicalisation process has been presented earlier in this section. Regardless of your standpoint, many type of intervention programmes are being carried out in many different countries. In SAFIRE we inventoried intervention programmes from countries both inside and outside the EU (WP 2.3). Basically, there are two types of interventions: preventative, aimed at making sure radicalisation does not occur; and restorative/suppressive, aimed at making sure radicalisation does not continue.
In work described earlier we identified factors contributing to the radicalisation of know terrorists. These factors were applied to the interventions: which factors were most often addressed by the interventions we inventoried? The most common were:

• Experience of political and personal events
• Proneness to violence
• Political activism
• Social influence
• Search for meaning

Notable is that the first three were also found in the earlier analysis on factors contributing to radicalisation. However, the final two (social influence and search for meaning) were not found in the earlier list. What this means is that, whereas there is some overlap between the factors that contribute to known terrorists’ radicalisation and the factors addressed by interventions, the overlap is not 100% complete. The factors indoctrination and inability to affect political change are not addressed by the interventions we inventoried, though they do play an important role in the radicalisation process.
In order to draw conclusions about characteristics of effective interventions, we had frontline workers evaluate the inventoried interventions (WP 3.1). The findings showed that effective interventions focus on the internal psychological experience of the participant, in terms of creating positive identity or reducing negative emotions. Building trust between the intervention participant and the practitioner is also key to success. The most successful situation is when the practitioner is themself an ex-radical.
Less effective interventions focus on the individual’s relationship with the external world and concepts like acceptance of authority.


A large part of SAFIRE was the longitudinal study, which tested hypotheses from different areas of the project in a field setting, and an interview study with ex-radicals to find out more about their narratives: why do they join radical groups and why do they leave (both WP5).

Longitudinal study

We collected data using interviews and questionnaires from participants and trainers of the DIAMANT programme, a preventive programme developed and carried out by SAFIRE partner SIPI, over the course of six months. Data were collected from two DIAMANT programmes. Participants in DIAMANT are young adults, usually with dual ethnicity, who have shown to be at risk of becoming disconnected to society through problems with school, work, finances or with the law. The goals of DIAMANT are to increase participants’ resilience, teach them to cope with different identities and contexts, and to keep participants connected to mainstream society.
The results are quite extensive and explored in detail in D5.1. Here we present the most important overall conclusions.
DIAMANT resulted in:

• More empathy towards the out-group
• Increased self-esteem and agency
• Less social disconnectedness
• Improved school attendance
• Higher employment

Taken together, we conclude that a system-based approach involving parents, school, and social workers helps to create a strong supportive network. This, in turn, helps keep lines of communication with young adults open when there are problems, which helps avoid isolation and disassociation with mainstream society.

Interview study

In an interview study we spoke with former right-wing extremists about their lives before during and after membership in the extreme right-wing group.

Why join? People tended to join the group in their teenage years. Often they were looking for an identity, a group to belong to. Joining the group was often considered a means to an end: being a member meant having friends or somewhere to go. In many instances, people were looking for like-minded individuals, who would embrace their ideas and philosophy. In many cases, there was a break from society in the form of lack of trust in authorities (e.g. police, government, teachers, parents) and a (perceived) threat from an out-group, such as the threat of losing a job. Often there is a trigger that eventually cinches the deal: an event in one’s personal life that changes thinking to doing. This is often the point that an individual will actually seek out or join an extremist group.

Why leave? Interviewees reported being members of the extremist groups for quite a long time: usually for at least six years. Eventually, however, they left the group. This could have been for a number of reasons. The interviewees mentioned bad group functioning (e.g. bad leadership), bad behaviour by members (e.g. fights, teasing, playing tricks on each other), fragmentation of the group (i.e. the emergence of factions), or incongruence with one’s own ideology. Importantly, again a trigger event in one’s own personal life was often the reason to eventually actually leave the group.


In the end, we have four main conclusions from our work.

1. All the work we have done point to the fact that radicalisation is an extreme version of a basically normal process. We may all be radicals at some level, the question is more whether or not our radical ideas can exist peacefully side-by-side with mainstream society. In cases in which peaceful coexistence is not possible, society may choose to intervene in the process. What we find, however, is that what is needed for intervention is not so different from what is needed in the case of other types of deviation from the social mean. Effective programmes are likely to resemble those used to deal with other issues.
2. Radicals are characterised by a lack of trust: in themselves, in the people around them and in society and authorities. Regaining this trust is key to successfully re-engaging with society. This is more likely to be achieved, however, by addressing issues of identity, self-esteem, self-respect, self-confidence and how to deal with negative situations and emotions, rather than by addressing the justness of those who should be trusted (e.g. restoring trust in authorities). Again, a lack of trust may not be particularly unique to radical individuals. Others who do not perceive a connection with society may experience this as well, such as the chronically unemployed, truants, people with problems with the law or even most ‘normal’ adolescents.
3. So why then, do some people become violent radicals while others do not? The path an individual take in their life is the result of the interaction between who they are and their context. Each person is different and each person perceives their environment differently. This can explain why two people from the same environment can have completely different narratives. And why one may turn to an extremist group to deal with their situation, while the other does not.
4. In theory it is both possible and ethically acceptable to intervene in the radicalisation process. However, in practice, the permissibility of intervening is a matter of cultural norms and law. Furthermore, one size does not fit all. Not all interventions work for everyone and in all situations. For this reason it is important to take into account individuals’ needs and narrative.
Potential Impact:
In SAFIRE particular attention has been paid to identifying the most important impacts of the work. D6.1 From Radicalisation Analysis to Deradicalisation: Policy and Field Recommendations is dedicated to identifying the most important impacts. In this section, we present a summary taken from Deliverable 6.1 in addition to discussion our dissemination and possibilities for exploitation.

1. Ethical debate
Due to the ethically sensitive nature of the topics of radicalisation and terrorism, specifically of non-violent radicalisation, SAFIRE came very close to not being carried out. Rightfully, there were concerns about the justification of research that is directly concerned with citizen’s right to free thought and speech and that looks at how to address citizens whose thoughts and speech are a potential threat to European society. In addition, there were concerns about carrying out research with people (potentially) engaged in varying degrees of radicalisation: how to avoid pitfalls of stereotyping, maintaining voluntary participation and true anonymity, and respecting their civil liberties while researching a very personal issue. Fortunately for us, the project was funded and, fortunately for European society, yielded valuable insights into how to deal with radicalisation and terrorism in democratic societies on one hand, and how to carry out research on this sensitive topic, on the other.

Radicalisation in European society
The threat violent radicalisation and terrorism pose in Europe is clearly part of the political and research agenda at the European level. However, the way different European countries approach this topic is quite different. Some countries, such as the Netherlands or Germany employ a more preventive approach in which it is acceptable in the public debate to discuss non-violent or pre-violent radicalisation in terms of its place in a peaceful democratic society. Moreover there is room for an open dialogue about the degree to which its evolution towards violence should be prevented: let’s say a nip-it-in-the-bud strategy. On the other end of the spectrum, we find countries such as France and Portugal where the right to think, say and do pretty much anything you want as long it is not illegal is inalienable. In such countries, radicalisation in terms of either its extra-legal characteristics or the threat it potentially poses to society is not open for debate, and thus for all intents and purposes, is a non-issue.
SAFIRE is not here to make value judgments about which approach is better or worse. However, we have concerned ourselves with the ethical and legal justification for intervening in the radicalisation process. Legally speaking, terrorism can be tried in court if terrorist intent can be proven. If it cannot be proven, then ordinary criminal liability is applied. Radicalisation, however, is not legally defined and therefore does not result in criminal liability. In short, in Europe, terrorism is a criminal offence, radicalisation is not.
So, legally speaking, there is no basis upon which to intervene in non-violent radicalisation. But is there an ethical basis? According to virtue ethics, individuals have the right – if not the duty – to live the good life. Though what the good life entails is unique for each person, what is common to all individuals is that in order to discover their own good life, they need to engage in critical thought and debate with others about what is best for them. In order to accomplish this, people need the skills for critical thought and a social environment that stimulates and supports debate. Seen in this light, discussion about radicalisation and preventively intervening in the radicalisation process is justified if the goal is to promote the quest for the good life. Note two things, however. First, this type of intervention is not only applicable for individuals who are (at risk of) radicalising, but for all individuals, particularly in their adolescence. Second, it is possible – and ethically acceptable – if after this process of soul searching, an individual’s conclusion is that for them the good life means being a violent radical.
One of the biggest obstacles to policy and debate about radicalisation and terrorism at the EU level is the difference in the way different EU countries define and delineate the issues. Only when these differences are harmonised in a way that makes effective debate possible, will progress be made. SAFIRE contributes to this process by making the differences explicit and by placing radicalisation in a broader ethical context.

Research on radicalisation
Aside from the inherent politically and socially sensitive nature of the debate on radicalisation, researching the process of radicalisation and radical individuals has its own risks and challenges. In SAFIRE we addressed these head-on in order to make steps towards developing guidelines to how to research such topics. We have developed working processes for three issues.
1. Direct contact with research participants. Researchers should communicate directly with radicals, rather than using a proxy, such as a social worker, to collect data. First, by doing so, data will be more direct and hence more reliable. Second, a person’s autonomy is better respected when they themselves describe their life and opinions, rather than when the information is gained through a proxy. Third, asking frontline workers about the people they work with could constitute a breach of the trust relationship.
2. Informed consent. In our work, many radicals did not want to sign the informed consent form, as this in and of itself formed a threat to their anonymity. To get around this, we developed an indirect procedure for obtaining informed consent, in which a trusted third party acts as a go-between.
3. Minors. An interesting age group in the study of radicalisation is between 12 and 18 years of age. However, including minors under the age of 16 raises serious issues regarding their ability to understand what is being asked of them and therefore be able to give consent. In SAFIRE, we argue for the use of participants with a minimum age of 16. For participants under 18, parental consent is still required. Due to the fact that radicalisation starts in adolescence, it is important to also study individuals younger than 16. For this, however, a suitable protocol for recruiting, informing and studying this group needs to be developed.

2. Radicalisation as a normal process
In the debate about radicalisation, one of the pitfalls is to consider evolution towards violent radicalisation something unusual, mysterious or even abnormal. The radicalisation process has a comparable aura to humans’ willingness to follow unconscionable orders, which Stanley Milgram investigated in his obedience studies in the 1950s and 60s: this is something that other people do, it would never happen to me. What we concluded after the three and a half years of SAFIRE is that the radicalisation process is not all that abnormal at all. In fact, in many ways it resembles the way many adolescents develop (most radicalisation occurs during adolescence), just in a particular and sometimes extreme form. The implications of this is specifically in regards to how to address radicalisation. If the radicalisation process is “normal” development gone awry, then principles of “normal” interventions apply to effectively curb this process.
This is a significant conclusion in that up until now, many interventions focussed on radicalisation were specifically geared towards radicals as a specific and unique group. The conclusion we draw, however, is that this is not necessary. Applying normal social work practices to radicals should be at least as effective.

3. Understanding radicalisation
Much research in the area of radicalisation looks at the factors that contribute to the radicalisation process either at the individual or at the group level. In SAFIRE we examined these factors in three ways in order to achieve a more comprehensive overview of the big picture.

Approach 1:
This focuses on representing the complexity of the radicalisation process by developing a methodology that combines and analyses the different causes and characteristics of radicals as well as adequate types of interventions. This approach resulted in a framework consisting of three types of factors that influence the progression of non-violent radicalisation to violent radicalisation and terrorism.
• Background factors: factors contributing to a period in which an individual may already be radicalised, but in which there is a lack of observable or reported internal consideration for criminality and/or radicalisation;
• Proximate factors: factors influencing an individual’s consideration to engage in violence or terrorism to further their radical objectives;
• Immediate factors: factors influencing an individual who is ready to commit a violent or terrorist act.

Approach 2:
This approach aims at developing a ‘reverse methodology’ of violent radical groups and individuals, by focusing on their operational and organisational aspects and thus providing insights into possible end results of the radicalisation process. Whereas Approach 1 focuses on factors that influence radicalisation, Approach 2 focuses on characteristics of radical groups and individuals. Importantly, this second approach explicitly looks at radical groups from a perspective that does not include the group’s ideology. This is in order to ascertain if it is possible to cluster groups and individuals on dimensions unrelated to ideology, such as leadership structure or capabilities. The work indicates that other dimensions, such as command and control structure, may be equally as meaningful as ideology when it comes to characterising and intervening in (violent) radicalisation.
This work indicates that to develop strategies to counter a radical group, focus on the operational characteristics of that group. In addition, an over-focus on ideology to the exclusion of other characteristics and behaviour, will produce a narrow and incomplete picture.

Approach 3:
This approach looks at radicalisation from a social network perspective. In this work, SAFIRE inventoried factors related to radicalisation on the individual, group and society levels. Sources were gleaned from many disciplines, such as psychology, political science, cultural science and security science. The relationships between the various factors were defined in order to gain better insight into 1) which factors influence each other most strongly and 2) which factors can be actively influenced to have a positive effect on the radicalisation process at any of the three levels (individual, group or society). These factors are referred to as actionable. The resulting network model is not meant to present one coherent picture of radicalisation, but rather, can be used to better understand specific elements of radicalisation as needed – for instance, the relationship between an individual’s psychological state and socio-economic status on one hand, and their sensitivity to radicalisation on the other. The impact of such a network-based approach is on the one hand that some factors related to radicalisation are actionable, others are not. Successful radicalisation policy depends on recognising and utilising the most promising actionable factor, given a particular context. In addition, as comprehensive a model of radicalisation as possible – which can be extended and modified as needed – is essential to improving our understanding of radicalisation, both at the conceptual level and at the practical level for use in the field.

These three complementary research approaches underline the extreme complexity of undertaking a comprehensive analysis of violent radicalisation. By taking a multimodal and holistic approach to radicalisation, in which the different modes partly rely on complementary debated views or assumptions, the findings of SAFIRE gain in robustness. It also illustrates that, even on a fundamental research level, underlying assumptions, cultural background or diverging security perceptions can have strong consequences on the way research protocols are defined.

4. Intervention effectiveness
Various countries carry out intervention programmes in order to halt, curb, or reverse radicalisation at various stages of extremeness and violence. However, there is little information regarding what works and what doesn’t. In this light, carrying out evaluation studies is essential to effective resource allocation and policy development. Furthermore, radicalisation research is hampered by the lack of a coherent research paradigm or protocol that can be used to study this phenomenon.
SAFIRE provides the community with guidelines for how to carry out studies on radicalisation. These are useful for both effectiveness studies, whose goal is to assess the effects and effectiveness of intervention programmes primarily for applied policy goals, and for scientific research, whose goal is to increase our basic understanding of the radicalisation process and the individuals who engage in (violent) radicalisation.
These guidelines include
• Recommendations for recruiting and selecting participants
• Protocol for indirect informed consent
• Research paradigm
• Operationalisation of effectiveness is measurable concepts
• Interview protocols
• Questionnaires
• Analysis methods

5. Economic impact
At this time, it is very difficult to say what the economic impact of SAFIRE is or will be. SAFIRE did not develop a tangible product that can be marketed for the purpose of economic gain. Instead, SAFIRE developed knowledge of the radicalisation process that can be used to better understand, prevent, or address the threat it poses to European society. The impact of this in a broad sense, can be a reduction of the threat of a terrorist attack and the financial, emotional and material damage such an attack causes. How big this reduction is, however, is impossible to say.
What we can say is that by using the insights developed in SAFIRE, it is possible to more effectively develop, carry out and evaluate policies and deradicalisation intervention programmes. This improves the quality of cost-benefit analyses and allows for a more effective and focussed allocation of funds.

6. Dissemination
Dissemination has been hampered by the fact that the first two reports (D2.1 and D4.1) were restricted. In consultation with DG ENTR, we agreed to develop two extensive public summaries of the restricted reports. In addition, we agreed on a distribution protocol for the complete restricted reports, which would allow individuals who can benefit from the details of the work to have access to it. This protocol is described in the fourth amendment.
Dissemination activities that have taken place are:
• We have produced three newsletters
• We have set up a LinkedIn discussion group
• We have produced a project flyer
• The website is updated regularly with new information
• We developed a second website to showcase the project results
• In June 2012 we held an expert meeting in Amsterdam, NL to obtain external feedback on our work.
• We held two expert meetings held in relation to WP4
• We held a final event in November 2013
• Various partners have given presentations and engaged in other dissemination activities.

7. Exploitation
SAFIRE was, from the start, a project designed to improve our knowledge of the radicalisation process. In this section, we look at different ways this knowledge can be exploited.

The academic partners in SAFIRE will primarily view the exploitation of the SAFIRE results in light of how they can improve their or their university’s track record and the services, mainly educational, offered by the institution. For these partners, the most important exploitation opportunities are peer-reviewed publications. During the project the publishing of peer-reviewed papers was hampered by the restricted nature of the work and the deliverables. To an extent, this was solved by drafting extensive public summaries and a protocol for distribution of the full restricted reports. However, for the academic partners, nothing replaces the need for peer-reviewed publications. This can only be resolved by writing papers now based on the unrestricted material and waiting for the restriction to expire in two years’ time (see GA amendment approved on 11 July 2013). Both options are sub-optimal, but there are few alternatives.
Regardless of the restriction of the actual publications, SAFIRE has greatly increased our understanding of the radicalisation process. This knowledge, while we are not able to release much of it in any widespread way at the moment, can be used as a basis for further research in this area. As this research is the core business of universities, this is a valuable exploitation opportunity for the SAFIRE results. In a specific initiative to build on the SAFIRE results, Dr. Bertjan Doosje, who was one of the prominent university professors in SAFIRE, has been named full professor, partly as a result of the work in SAFIRE. He holds the FORUM-Frank Buijs Chair on Radicalization Studies at the University of Amsterdam.
Finally, in addition to research, universities’ core business is providing education. There are concrete initiatives at the University of Coimbra, University of Amsterdam, and the University of Applied Sciences in Utrecht to incorporate SAFIRE methods and results into their educational curriculum. Work in SAFIRE is being used in developing lectures, interactive workshops and as input for internships.

Track record
As a precursor to commercial exploitation, it is important that organisations such as the CROs in SAFIRE use the knowledge to develop and solidify their track record in the area of radicalisation. This improved their ability to position themselves as consultants and advisors for national and European authorities and government bodies.
For organisations who operate closer to the field, such as SIPI and Forum, SAFIRE provides an opportunity to demonstrate the effectiveness of programmes and policy, which can impact future policy directions and funding allocation. This goes directly to their core business, and provides such organisations leverage to help set agendas.

Commercial exploitation
SAFIRE was not designed as a project that would develop a ‘product’ that can be commercially marketed. Commercial exploitation of the work in SAFIRE is mainly foreseen to be in the form of follow-up projects from relevant authorities. At the moment, TNO is in contact with the US authorities in the Netherlands and the Ministry of Security and Justice to discuss their goals and how TNO can potentially aid them in achieving these goals.
Alternatively, ISCA’s primary business comes from providing trainings for security personnel in detection and action regarding those individuals who pose a security threat. ISCA can use the work in SAFIRE to improve or augment existing trainings or develop new trainings specifically for dealing with radicalisation.
Partners active in the area of consulting and research specifically in the area of security, such as FRS and CEIS can apply the work from SAFIRE as background knowledge with which to better serve the needs of their clients. In this way, for example, FRS has organised a training for the Gendarmerie based on work in SAFIRE.
List of Websites:
public project website
public website showcasing the project results

contact details:
Heather Griffioen (coordinator)
+31 88 866 5931 (T)
+31 6 2246 1065 (M)